Colorectal cancer-related mortality is most strongly related to the extent of tumor cell invasion and metastasis at diagnosis. While patients with mucinous colon cancer appear to have a poor prognosis, it is not yet known which structures in colon cancer mucin are most important to the biological behavior of these cells, or which of the many steps in the metastasis cascade depend on specific alterations in secreted or cell-surface glycoproteins. The long term objective is to further elucidate how the cell surface and secreted glycoproteins of colon cancer cells influence their metastatic capacity. It is proposed that sialomucins are preferentially secreted by colon cancer cells with high metastatic ability, and that sialylated carbohydrate structures are functionally important portions of the mucin molecule in metastasis. Altered sialylation of metastatic colon cancer cells leads to cell-cell and cell- substratum interactions which determine the ability of these cells to metastasize. The aims of the proposal are to determine the relationship between production of sialomucin by colon cancer cells and metastatic capacity, to determine how the type and amount of apomucin affects metastatic capacity, and to determine the functional significance of mucins in individual stages of metastasis. Animal models developed by the PI have been used to select human colon cancer cells which differ in their capacity to spontaneously metastasize from cecum to liver, and to colonize the liver after splenic-portal injection. Mucins from these cells and from surgical specimens will be purified and characterized with relation to sialylated carbohydrate structures. Conversely, the metastatic ability of cells which produce mucins that vary in their expression of sialylated antigens will be tested in in vivo models of metastasis. Which sialic acid linkage and which structures are functionally important in metastasis will be addressed. Human mucins are encoded by seven or more distinct mucin genes, at least two of which are expressed in the colon. The relationship between type of apomucin and colon cancer metastasis will be assessed, including the effect of specific inhibition of mucin gene expression on experimental metastasis. How colon cancer mucins and associated carbohydrate structures affect tumor cell interactions with basement membranes, endothelial structures and endogenous carbohydrate binding proteins will be evaluated to clarify the role of mucin in individual stages of metastasis. With further understanding of these relationships, it may eventually be possible to identify patients with a poor prognosis and design specific therapies targeted against metastatic disease.